Near miss with 5 airliners waiting for T/O on taxiway "C" in SFO!

[RAT 5]

One can surmise that reducing to a single multi-tasking ATC controller at supposedly quiet times is purely a cost saving issue. As has been mentioned on many issues over the past decades there will a point where cost saving starts to impact on safety. The skill in risk assessment application is balancing the cost savings with still providing an acceptable of level of safety. We've all noticed over the past couple of decades how many items have been dumbed down: e.g. reduced reporting times with self briefing; the self-help printing of nav' plans, NOTAMS, weather etc., the removal of crew apron transport = longer times from crew room to a/c all causing less time for in-depth flight preparation, reduced turn-round times = expedited flight preparation for following sectors, reduction in number of licensed engineers, ATC controllers on multi-frequencies, reduced pilot training and SOP's that dilute pilot handling skills, reduced experience for command upgrade, etc. etc.
I wonder if the industry is close to tipping points in some areas. I wonder if TEM & risk assessment techniques are being applied to supporting links in the chain in the same way we are taught to apply them in the cockpit. I wonder if the bean counters just go-ahead with their scalpel un-supervised until it becomes obvious that there is too much bleeding as the fat has been cut to the quick?

[underfire]

Searidge has a tracking capability in their software (not sure about MPH!?!) This does warn if the ac is not on the runway.

[ZOOKER]

ATCOs recording ATIS broadcasts?

[Bedder believeit]

In Hong Kong where I last worked in the tower (2 years ago) the tower assistant prepared the basic ATIS info which is mainly electronically generated, but the tower controller had to authenticate the information contained therein before broadcasting the ATIS. This was critical when undergoing runway changes and/or trying to keep ahead of rapidly changing weather conditions. As yet I doubt that there's an ATIS mechanism that is able to anticipate all of the parameters that a live breathing controller is able to forecast.
As for the controller's ability to accurately assess runway alignment for approaching aircraft, in Australia, Dubai and Hong Kong where I worked, the tower controller had a readout of the terminal radar display that indicated where the landing aircraft were tracking. I don't know what happens in the States, but I would have to guess they have something similar.
Just remember, that for every radio transmission that the system records as ATC radio "chatter" is probably about 20% of the controller's workload, particularly when it's busy.
At 2 in the morning it can get pretty demanding.

[Obese]

It is difficult to believe that Canadian duty time rules are really intended to promote aviation safety, but, nevertheless, as a paying airline passenger, I would minimally expect aircrew to listen to destination ATIS prior to descent, especially on an international night flight. Kudos to the Filipino captain for apparently being the first on the ground to activate landing lights. It may have been the brick on the side of the head that Abbott and Costello needed to prompt them to initiate the go around.

[Ian W]

Originally Posted by peekay4

Why should there be a limit?

The controller was already there, and in his own judgement, if he wasn't doing other tasks he might have been able to detect that the aircraft was lined up on the wrong runway.

Why not support him by:

1. Not putting him on a schedule resulting in 2 hours of sleep in 24 hours on the backside of his circadian rhythm
2. Making available a second controller to share his workload, per FAA guidelines
3. Giving him time & space to observe critical phases of flight, instead of doing admin tasks
4. Providing additional tools / technology to detect wrong runway incursions

49 people perished on that crash. If another set of eyeballs had a chance to prevent the accident, why not take advantage of that?

I don't know how much time you have spent in the visual control tower at night, but it is not easy sometimes to even see an aircraft let alone be certain where it actually is on closely located runways.

But let's assume that the controller was actually supported and only doing visual control. The aircraft taxies to the far end of the airport and calls for takeoff in a position that could be either runway an unlit extra narrow GA runway with a divergent heading or a fully lit standard runway.
The aircraft is cleared takeoff.

I am watching carefully and it seems to be slower than normal on the take off as its angle is not changing so fast after about 15 seconds I think that the aircraft can't be on the active runway.... it could be on the GA runway.

Now - the controller has a VERY difficult choice - one that I am sure you have never had to make.
IFF the controller tells the crew you are on the wrong runway abort takeoff - the aircraft is likely to overrun the end of the runway with a similar result to the crash as the cancel clearance call could be after the aircraft was committed- and the peanut gallery will be saying that they could have become airborne - the controller caused the crash
IFF the controller says nothing as the perception may be wrong - then the controller is in a worse position than now - as the crash MIGHT have been prevented if the cancel clearance call was made before the aircraft was committed (when is that on a 4000ft runway that is already too short?) and the peanut gallery would be saying the controller caused the crash

So the NTSB asked an almost impossible to answer question - could you have prevented the crash if you had been watching - and the controller correctly said - he MIGHT have been able to.

I doubt if anyone in the NTSB actually put a CRJ out on the GA runway at night to see when it would be apparent to a controller that the aircraft was actually on the GA runway not the main runway. Then with that point decide if the aircraft could have been saved after the slight delay for controller 'surprise' that the aircraft was not on 'the' runway.


This is like the Sully assessments that he should have turned back to LGA as he _might_ have made it. All very good in hindsight if every decision had been just right.

[PukinDog]

Too bad the CVR was conveniently overwritten. If it weren't, investigators would know if what should have occurred before this approach began actually occurred, including the crew using the primary, fundamental tools one does in order to plan/know what to expect; bookwork like we're paid to do, planning, and a proper briefing as dictated by Company SOPs. As it is, nobody knows if they followed any or conducted an approach briefing, or even if they used checklists appropriately or observed a sterile cockpit. We do know this, however;

The NOTAMS re the closure of 28L and 28Ls Approach Light System were both published. AC OPs Specs would dictate they were disseminated to the crew.

The ATIS received by the crew before the approach also notified them that 28L was closed and the Approach Light System for 28L was out of service. Company SOPs would dictate this be reviewed by the crew prior to the approach and ATC would expect acknowledgment it had been received.

The publications/Airport Diagram the crew had on hand showed the parallel runways with Approach Light Systems installed on each, and details or each installation plus runway lighting and PAPI. They also show Taxiway C running parallel to 28R.

The crew told the investigators that they thought 28L was 28R. That is admitting they didn't either read the NOTAMs or listen to the ATIS, or means they did but never talked about it/ignored what would could be an operational concern (runway closure).

It also means they were unaware of what they should be seeing for miles before they attempted to land on Taxiway C; an Approach Light System and a PAPI.

A proper approach briefing of any type identifies the primary guidance to be used, and most Company SOPs dictate their review and inclusion. This was a visual approach, at night, over water where any licensed pilot should be well-aware of possible black hole effect which makes the vertical guidance all the more critical to identify. Runway 28R has a diagramed and functioning Approach Light System (ALSF-II) stretching out into the water for lateral guidance. 28R has a PAPI located on the LH side of the runway for vertical guidance. These are ground-based, visual approach light systems that lead directly to the runway and Touchdown Zone of the correct runway. Those are the primary aids for the approach, not the FMS.

How hard is it to brief (and it makes no difference which FMS procedure got them to the point or what green taxiway lights look like from a distance) "the visual segment for 28R will be over water, we'll align with the ALSF-II and follow the PAPI located on the LH side"?

That's what a briefing is all about. To create a picture of what is known and printed right there in our faces about where we're going, and what to look for out the windscreen when the time for looking is at hand.

Briefings, SOPs, and checklists are there to backstop performance, especially when tired, etc. because they focus attention on where it needs to be focused. Even the well-rested pilot that does no bookwork, planning, or briefing will stuff things up a thousand times more often than a tired pilot that does all the above.

One can go around and around about FMS procedures, lack of EVS, "Tunneling" and circadian rhythms, but those aren't going to prevent stuff-ups of this nature if the crew didn't bother with fundamentals and basic adherence to procedures already in place.

Does anyone actually think a detailed, proper briefing that reviewed the primary guidance aids (ALSF-II and PAPI) to be followed for this visual approach was conducted in this case?

The statements by the crew and convenient overwriting of the CVR lead me to suppose this approach wasn't stuffed-up on short final, or 4 miles out due to an illusion, but long before when no attention to detail to what was published and/or briefed regarding the upcoming visual approach to the runway they were (supposed to be) landing on. If they had briefed the critical details of primary guidance aids just like everyone does for an ILS etc, this wouldn't have happened. The aids were right there, yet they never looked for them or (obviously) used them until the 2nd time around. Bear in mind that 2 crew with separate sets of eyeballs missed the same things.

I don't believe there's any big mystery or highly-unusual Human Factors, super-illusion boogeyman at work here. We deal with many all the time, usually by preventative measures. Until it can be shown that this crew followed those time-proven, fundamentals that prevent this sort of thing from happening I'm not going to assume it.

[peekay4]

I don't know how much time you have spent in the visual control tower at night, but it is not easy sometimes to even see an aircraft let alone be certain where it actually is on closely located runways.

True, but there have been two other instances at LEX where a jet incorrectly lined up on 26 instead of 22, and in both cases ATC detected and alerted the flight crews.

In fact one of the incidents happened shortly after the Comair crash in similar conditions. A small Lear 45 was taking off at night and turned to the wrong runway. An alert controller saw it and immediately cancelled the Lear's takeoff clearance.

As reported by the controller:

ASRS ACN: 722668

Date : 200701
Local Time Of Day : 1801-2400
Locale Reference.Airport : LEX.Airport
Flight Conditions : VMC
Light : Night

Narrative (LEX.Tower):

LJ45 TURNED ONTO RWY 26 AFTER GIVEN A TKOF CLRNC ON RWY 22. AN IMMEDIATE CORRECTION WAS MADE BY ALERTING OF THE WRONG RWY -- BEFORE TKOF ROLL BEGAN. TO PREVENT A RECURRENCE SUCH AS THIS ONE, INCLUDE IN THE TKOF CLRNC FOR RWY 22 TO CROSS RWY 26.

Synopsis

LEX CTLR DESCRIBED AN ATTEMPT BY AN LJ45 FLT CREW TO TAKE OFF ON THE WRONG RWY AS THEY TURNED ONTO RWY 26 WHEN RWY 22 WAS ISSUED.

[slack]

With regards to post 789 by Ian I couldn't agree more. Forget the fancy FMS stuff. Using info from notams/atis and their God given eyes (2 sets ) in severe clear wx was all that was required for a routine successful landing.
P.S. I am a retired ( thank God ) a320 pilot

[llondel]

Originally Posted by PukinDog

The NOTAMS re the closure of 28L and 28Ls Approach Light System were both published. AC OPs Specs would dictate they were disseminated to the crew.

The ATIS received by the crew before the approach also notified them that 28L was closed and the Approach Light System for 28L was out of service. Company SOPs would dictate this be reviewed by the crew prior to the approach and ATC would expect acknowledgment it had been received.

The publications/Airport Diagram the crew had on hand showed the parallel runways with Approach Light Systems installed on each, and details or each installation plus runway lighting and PAPI. They also show Taxiway C running parallel to 28R.

The crew told the investigators that they thought 28L was 28R. That is admitting they didn't either read the NOTAMs or listen to the ATIS, or means they did but never talked about it/ignored what would could be an operational concern (runway closure).

This one got me thinking. Unlikely, but assume they'd made a mistake and misheard the ATIS that 28L was closed and 28R approach lighting was out. (I assume 'approach lighting' means the stuff sticking out into the bay). Under this scenario you'd expect them to do exactly what they did because they'd be lining up on the tarmac lights to the right of the working approach set. Of course, this immediately begs the question of why the approach lighting would be set up in this configuration.

[Akali Dal]

Originally Posted by PukinDog

Briefings, SOPs, and checklists are there to backstop performance, especially when tired, etc. because they focus attention on where it needs to be focused. Even the well-rested pilot that does no bookwork, planning, or briefing will stuff things up a thousand times more often than a tired pilot that does all the above.

One can go around and around about FMS procedures, lack of EVS, "Tunneling" and circadian rhythms, but those aren't going to prevent stuff-ups of this nature if the crew didn't bother with fundamentals and basic adherence to procedures already in place.

Does anyone actually think a detailed, proper briefing that reviewed the primary guidance aids (ALSF-II and PAPI) to be followed for this visual approach was conducted in this case?
.

But that cannot be. The pilots of this flight are from a first world nation with a super proud tradition of meticulous training and water tight integrity. I am sure the CVR wasn't deliberately meant to be overwritten! They have sound fundamentals and very strict checks and balances. The runway layout, the unnecessary runway closures and unhighlighted notams, the ATC, the tiring schedule, the circadian rhythms, and a whole lot of other contributing factors are largely to blame.

[underfire]

But that cannot be. The pilots of this flight are from a first world nation with a super proud tradition of meticulous training and water tight integrity.

No, they are from Canada.

They have sound fundamentals and very strict checks and balances. The runway layout, the unnecessary runway closures and unhighlighted notams, the ATC, the tiring schedule, the circadian rhythms, and a whole lot of other contributing factors are largely to blame.

Yes, this will likely be the findings....

As noted in other posts by ATC folks...it happens all the time. What is done about it....nothing. Nothing to see here , move along.

Blame, blame, blame...take responsibilty, NO WAY.

The statements by the crew and convenient overwriting of the CVR lead me to suppose this approach wasn't stuffed-up on short final, or 4 miles out due to an illusion, but long before when no attention to detail to what was published and/or briefed regarding the upcoming visual approach to the runway they were (supposed to be) landing on. If they had briefed the critical details of primary guidance aids just like everyone does for an ILS etc, this wouldn't have happened.

Typically, when the FMS Bridge visual or any of the visuals are run on 28R, 28L cannot be used due to the close prox. In that note, it really does not matter that 28L was closed and there was a big white X at threshold. When 28R visual is in use, the lights for 28L would not be there anyways.....forget about all this talk about the 28L lights were not on, so mistook 28R lights for 28L and offset, into the bay......they would never had been on.

[peekay4]

Typically, when the FMS Bridge visual or any of the visuals ar run on 28R, 28L cannot be used due to the close prox.

That's not correct. At SFO both runways often operate in parallel during visual approaches, e.g., while 28R is on the FMS / Quite Bridge visual, aircraft on 27L might be on the Tipp Toe visual (along the ILS 27L localizer).

[underfire]

no, they do not and cannot.

SOIA is not authorized visual. (the hint here is the title...Simultaneous Offset Instrument Approach)

27L might be on the Tipp Toe visual (along the ILS 27L localizer).

oh, you must be Canadien...landing on 27L...blame it on map shift, or circadien cycles. Good Luck! Try not to hit anything.

[peekay4]

no, they do not and cannot.

SOIA is not authorized visual.

I've been saying many times now, a VISUAL APPROACH IS NOT AN INSTRUMENT APPROACH PROCEDURE.

Parallel visual approaches have nothing to do with SOIA.

Note on the FMS Bridge Visual 28R:

Closely-spaced parallel visual approaches may be in progress to Rwy 28L utilizing ISFO ILS 28L localizer.

Notes from TIPP TOE VISUAL if you're shooting the visual on 28L:

CAUTION: AIRCRAFT OPERATING WITHIN 500' TO PARALLEL RWY- POSSIBLE WAKE TURBULENCE

NOTE: Closely spaced parallel visual approaches may be in progress to Runway 28R utilizing the SFO R-095. In the event of a go-around on Runway 28L, turn left heading 265°, or on Runway 28R, turn right heading 310°, climb and maintain 3000, or as directed by Air Traffic Control.

If criticizing pilots, at least try to get basic facts right.

[Airbubba]

Originally Posted by underfire

In that note, it really does not matter that 28L was closed and there was a big white X at threshold. When 28R visual is in use, the lights for 28L would not be there anyways.....forget about all this talk about the 28L lights were not on, so mistook 28R lights for 28L and offset, into the bay......they would never had been on.

So, now you claim that the lights to 28L are turned off whenever there are visual approaches to 28R?

Lad, something tells me you don't know much about flying a plane.

[underfire]

No , that is not what I stated.

Have you flown into SFO at night on the 28R visual approach?

This is exactly what it looks like.



With all of your experience, would you consider the approach lighting for 28L on? Would you line up with taxiway C instead of the approach cross?

[peekay4]

Have you flown into SFO at night on the 28R visual approach?

This is exactly what it looks like.

No, that's only what it looked like during that particular approach. You're being misled by what you see (only) online, and making wrong conclusions based on that.

Here's a frame from a night visual approach to 28R while 28L was also operating. You can see the 28L lights (MALSR) are on:

[Ian W]

Originally Posted by peekay4

True, but there have been two other instances at LEX where a jet incorrectly lined up on 26 instead of 22, and in both cases ATC detected and alerted the flight crews.

In fact one of the incidents happened shortly after the Comair crash in similar conditions. A small Lear 45 was taking off at night and turned to the wrong runway. An alert controller saw it and immediately cancelled the Lear's takeoff clearance.

As reported by the controller:

Well the one shortly after the crash does not surprise me one bit all the controllers will be on edge about crews not able to check runway heading and be using binoculars to check every aircraft departure. Although I am surprised the crew was not more aware.

I cannot comment on the similarity or not of the previous incident. The fact still remains that visually assessing from a distant control tower where an aircraft is lined up at night when the angle subtended between the correct and incorrect positions is very small is not guaranteed.

[peekay4]

Yeah. "Shortly after" is relative here. The incident that preceded the crash happened in 1993 (13 years prior). The Lear incident happened just 4 months after the crash. I don't know if controllers would (still?) be using binoculars to check every single take off, months after the crash?

The fact still remains that visually assessing from a distant control tower where an aircraft is lined up at night when the angle subtended between the correct and incorrect positions is very small is not guaranteed.

Agreed.